INDEX

Explanations

terms related to cancer, specifically focusing on types of tumors and invasion

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 pelajaran

-0.43

 paille

-0.42

 autorytatywna

-0.41

 gyhoeddwyd

-0.40

insegna

-0.40

 plazos

-0.38

 rahasia

-0.38

 empleos

-0.38

 keter

-0.36

 prestasi

-0.36

POSITIVE LOGITS

 invasion

0.93

 invasions

0.83

 invade

0.82

 attack

0.80

 invaded

0.80

 invas

0.78

invasion

0.77

 inva

0.77

 Invasion

0.75

 invading

0.73

Activations Density 0.657%

terms related to cancer, specifically focusing on types of tumors and invasion

No Comments

No Known Activations